NOISE FROM GEOTHERMAL DRILLING

Transcription

1 NOISE FROM GEOTHERMAL DRILLING E. Patsa 1 and S. Zarrouk 2 1. Department of Mining Engineering, University of British Columbia, Canada; lena@mining.ubc.ca 2. Department of Engineering Science, University of Auckland, Auckland, New Zealand; s.zarrouk@auckland.ac.nz ABSTRACT High levels of noise are emitted during drilling and more specifically during the drilling of deep geothermal wells; this is due to the implementation of under-balance drilling, which minimizes reservoir damage. Impact assessment included measuring noise emitted from particular equipment on the drilling site and assessing the exposure of drilling personnel during regular activities. The main noise producers were identified as the generators, air compressors, the draw-works on the rig floor and the cementing unit. The personnel most exposed to noise were the roughnecks and drillers. An experiment comparing three different noise-cancelling materials was also conducted on a quiet site using a known noise source. From the three materials tested (specialty sound-cancelling sheeting, shipping containers and hay bales), the hay bales proved most effective, mainly due to their overall mass and total surface area. Based on the results of this study and existing Occupational Health & Safety (OHS) regulations, this work identifies primary high riskareas of noise exposure on a large geothermal drilling rig and gives recommendations on mitigation measures. Keywords: Geothermal drilling, Noise, Impact assessment, Mitigation measures 1 INTRODUCTION Drilling is an essential part of geothermal energy utilization, as it is the process through which the energy carrier be it either steam or brine buried deep underneath the Earth s surface is harvested for the production of electricity. Unlike oil and gas, geothermal development often takes place in close proximity to populated areas. Sustainable energy production requires continuous drilling activity, as fluid extraction and can result in declining flow rates and pressures from existing wells that need to be addressed through the drilling of makeup wells. Furthermore, future projections of electricity usage call for the creation of additional capacity and new development is always underway at operating geothermal fields. As drilling operations can be in close proximity to populated areas, it is highly important to assess and mitigate any associated impact to people and the environment. Noise is an important environmental risk and it can be substantial throughout the entire spectrum of drilling activity. The main objectives of this investigation are as follows: a. to identify the main sources of noise on an active geothermal drilling rig; b. to assess the extent of exposure for people working in a drilling rig; c. to evaluate the suitability of current noise mitigation measures (i.e. Personal Protection Equipment); d. to estimate the effectiveness of different sound cancelling materials used on a drilling rig; and e. to propose a suitable noise impact mitigation strategy. 2 BACKGROUND The simplest definition of noise is unwanted sound [1]. Sound is a succession of travelling pressure waves moving away from a source that carry energy, which is commonly measured as emission sound pressure level at a specific location in precisely defined acoustical conditions [2, 3]. Sound pressure level (SPL) is the local pressure deviation from a reference point, taken as the lowest sound intensity detectible by the human ear (~2e-5 N/m2) and is expressed in decibels (db) [3, 4]. SPL = 10log!"!"#$$%"#!"#"!!!"#"!"$%"!"#$$%"#! (1) Equation 2 calculates the hemispherical noise propagation from SPL readings around a source [2]: SPL = SWL 20 log r 8dB (2) SWL is the Sound Power Level and it is defined by equation 3 below. Here reference power = 10e-12 watts. SWL = 10log!" (!"#$%!"#$%) (!"#"!"$%"!"#$%) The human ear is sensitive to different frequencies. Sound frequency is defined as the number of like waves that pass by a fixed observation point per second and it has the unit of Hz or cycles per second. The lowest frequency that can be identified as sound by human beings is 20 Hz and the highest can be up to 20,000 Hz, (3)

2 depending on a person s age and health [2]. A-weighting is a curve relating relative SPL (in db) with frequency (in Hz) and it reflects the frequency sensitivity of the human ear at low SPLs. Sounds that are weighted with the A-curve are referred to as dba [5]. Exposure to noise is important as health consequences of elevated SPL exposure can be severe and include hearing impairment, hypertension, ischemic heart disease, annoyance and sleep disturbance [6, 7]. Hearing loss may be temporary or permanent. In general, it is believed that brief exposure to noise, causing significant temporary hearing loss or threshold shift, may lead to permanent hearing loss if the noise exposure is prolonged or recurring [7]. Noise levels are therefore closely monitored in the workplace under direct specifications national and regional Occupational Safety and Health (OSH) organizations [8]. SPL levels were measured directly in this study with the use of three separate instruments [9, 10, 11], which were used to record the following properties: a. Time-Average A-Weighted SPL (L Aeq [in db]); b. root mean square maximum/minimum A-frequency-weighted SPL (L max/min [in db]); c. peak values of SPL (SPL peak ) and d. amount of time (in %) of duration noise levels we above (n) db (t n ). 3 RIG NOISE SURVEYS The term Rig refers to a particular assembly of drilling equipment. The rig itself occupies a rectangular area 120 meters long by 60 meters wide. The exact placement of the each piece of equipment on this rectangle (setup) is important, as noise can be a combination of multiple sources. Setup can change, not only between deployments but also while one a single site, to accommodate the various requirements of drilling activity. Although an equipment list can change through the drilling operation and components grow old, malfunction and need to be replaced, the Rig Specification has been assumed constant for the duration and purposes of this short-term study. 3.1 Point Source Measurements The first objective of this study was to identify the main sources of noise from operating equipment on the rig. For this purpose, a noise survey was conducted during which major known noise generation components were targeted. Five distinct functional areas were identified on site: (A) the generator shack; (B) the air drilling equipment and office; (C) the mud pumps and tanks and associated components; (D) the drill floor with the Tesco unit and the drawworks engines; and (E) the cementers setup. The measurement locations were selected to represent areas were workers spend a considerable amount of time in; for example, during cementing, locations E1, E2 and E3 were the only ones selected as most activity is limited within these points (Figure 1). The survey itself consisted of sequential SLP measurements at each map location on the map. For a typical measurement, the equipment was placed roughly one meter away from the source and the microphone direction was noted [12]. Taking a measurement closer that one meter from the source was not deemed appropriate, as typical exposure would not have a person s ear in such close proximity to this type of noise source. SPL readings (L Aeq [db], Min [db SPL ] and Max [db SPL ]) were taken for a period of five minutes at a time. Concurrent readings of wind speed and direction were also taken, at approximately two meters above ground. It is worthwhile to note that for locations that were shielded from wind effects, a non-zero wind reading is mainly attributed to an exhaust from the targeted piece of equipment, while the primary direction of wind was in the majority of these cases parallel to the microphone. Fifty-eight readings in total were taken at the forty-five locations. Of the five functional areas identified in the preparation stage, four had a maximum value above 100 db. The maximum L Aeq measured in the fifth area, the mud tanks, was very close to the 100dB mark, at L Aeq = 99.4dB. In total, 83% of readings were above 80 db, 42% above 90 db and 14% above the 100 db mark (L Aeq ). These levels indicate the importance of ensuring all personnel are fully protected when working near these components. One of the measurements was taken facing upwards underneath the rig floor pointing at the bottom side of the drawworks (location D8). A second reading was also taken in front of the drawworks, which was lower than the maximum db. The higher reading was due to the driller cleaning the hole, described as lifting the drill bit about two meters above maximum depth and smoothing the hole created. The second reading was taken during more regular drilling activity, at the 1-meter mark, which resulted in a lower SPL level. With the use of Equation (2), L Aeq at the bottom of the drawworks was calculated to db (referenced back to one meter from the bottom surface) compared to 94.8 db in front of it. This is a clear indication that more noise is emitted from the bottom of the structure than the top.

3 Figure 1: Rig setup map, showing measurement locations and corresponding SPL levels. 3.2 Personal Exposure Surveys The second objective of this study was to evaluate the potential impact of noise from drilling activity on people working on site. For this purpose, rig personnel were fitted with a personal noise dosimeter device. Individual impact was assessed for the duration of a typical 12-hour shift. The exposure of personnel working at four key positions was monitored and included the Driller on the break, the Roughnecks on the floor, the Derrickman tending to the mud tanks and pump operation and the Air Drillers. The personal noise measuring device was fitted on the clothes of each wearer, in close proximity to the ear and free of any interference with the wearer s clothing. This was not always physically possible, due to the nature of the work involved and the type and amount of protective clothing required when working on site. Each position was monitored twice in a 24-hour period. Each wearer was also asked to record their activity during the shift in an effort to place the captured sound pressure levels into context. The results of the personal exposure surveys are summarized in Figure 3. In all, eleven sessions were fully recorded, ten of which yielding usable results. The main observation deriving from these findings is the fact that all of the positions monitored are exposed to high levels of noise. The peak readings can be misleading as they could have resulted from banging the tip of the microphone during the process of putting it on or by the occasional, playful shouting. The highest level of exposure was that of the Roughneck, which is attributed to the amount of time actually spent on the drilling floor, in close proximity to the drawworks. In comparison, the Driller is somewhat shielded by his location. 4 NOISE-CANCELLING MATERIAL EXPERIMENT In this part of the study, the noise cancelling effectiveness of different materials was evaluated, in an effort to select the most appropriate as part of an overall mitigation plan for managing drilling noise. A completed (and hence quiet) re-injection pad served as the location of the experiment. Three different barriers were constructed on the site and placed roughly 20 meters apart (Figure 2). Noise was generated using a portable air compressor mounted on a trailer. The actual noise came from the compressor s adjustable air nozzle, which was adjusted accordingly for maximum effect. The trailer was placed at three locations in front of the barrier, at 0.8 m, at 3 m and at 5 m. The first barrier was constructed of six Wave Bar sheets, a flexible noise barrier manufactured by Pyrotek Soundguard [13]. Each sheet was 6 meters long, 1 meter wide and 0.04 meters thick with a weight of 10

4 kg/m 2. The entire assembly, weighing an excess of 300 kg was mounted and secured on an inclining scaffold. The Wave Bar sheets were taped together at the seams to minimize noise leakage. Sand bags were used to seal the bottom side on the barrier. Two full sets of measurements were conducted (at 90 o and the maximum allowable inclination of 66 o ) based on the design of the scaffolding substructure. The second barrier consisted of two hollow, steel, shipping containers, each 2.9 m high, 2.4 m wide and 6 m long. The containers were placed one on top of the other to create a wall 5.8 meters high. The third and final barrier consisted of 15 rectangular hay bales, with rough dimensions 0.9 m high, 0.85 m wide and 2 m long. The main body of the hay-bale wall was constructed with twelve hay bales stacked four up and three across. Three hay bales were used on the front side of the wall to provide additional support to the wall body. The experiment was conducted four times in total: twice for the Wave Bar barrier (at 90 o and 66 o ), once in front of the container wall and once in front of hay-bale wall. For each test, the noise source was positioned at one of the three available locations. The compressor was turned on and engaged to produce noise. An initial reading in front of each barrier was taken to ensure the noise produced was consistent throughout the three experiments. The instrument assembly was placed at each marked measurement location behind the barrier and a five-minute SPL reading was taken (L Aeq [db], Min [db SPL ], Max [db SPL ]). The set of seven behind-the-barrier readings was repeated for the three source locations. An additional set of measurements was conducted without the use of a barrier. For this no-barrier test, only seven measurements were taken in front of the noise source. These readings were used as a comparison reference point for all barrier tests. Figure 2: Noise-cancelling-material experiment on-location setup. 4.1 Experimental results & discussion When the three materials are viewed in comparison with one another (Figure 5), it becomes evident that the denser, more absorptive hay-bale wall provided the greatest amount of noise protection. This wall had the additional competitive advantage of having the largest effective surface area, from the two bales placed in front of it to support the overall structure. In the case of the Wave Bar wall, the 66 o inclination exhibited the most amount of air leakage and noise interference due to the lack of insulation from the bottom of the wall, which was raised above the ground and held down with sand bags. By comparison, the 90 o inclination suffered from much less air (and hence noise) leakage, as the entire bottom of the structure was aligned with the ground surface. The measurements taken behind the Wave Bar Barrier at both inclinations (90 o and 66 o ) were also are contrasted, in an attempt to highlight the effects of varying the wall s angle of attack (Figure 5). There was no significant difference between the readings when viewed collectively. This is mainly attributed to the presence of interference from all sides, which affected the SPL values measured. Side

5 interference also affected the hay-bale and container-wall readings (Figure 4), although the effect was much less pronounced with the two latter walls, due to their overall mass. Figure 6 was drawn from two set of readings conducted without a barrier, one under conforming conditions (i.e. were wind speed <5 m/s) and one during which wind speeds exceeded 5 m/s. This comparison may indicate that the effects of the wind on the validity of the noise measurements are greatest closest to the wall and tend to dissipate as one moves away from the wall. 5 CONCLUSIONS A good set of spot readings was achieved that allowed for the identification of the principal noise producers on the rig. All measurements were completed during drilling and represent only a snapshot of the situation, as the procedure adopted did not account for the effects of attenuation, elevation and interference from neighbouring components. The spot measurements identified the generators, the ADA booster, the drawworks on the floor and the cementing unit as the highest noise emitters on the rig, at or above 100 db L Aeq. A personal exposure survey was also completed, which monitored the noise exposure of four rig positions. The roughneck position was identified as the one with the highest level of noise exposure, although the remaining positions were found to be at comparably high levels. A full set of experiments was conducted, resulting in a qualitative assessment of the noise-cancelling materials under investigation. The hay bale barrier was found to be the most effective in lowering SPL levels within the scope of these tests. 6 RECOMMENDATIONS To effectively control and mitigate noise, it is crucial to gain a full understanding of the particular conditions that combine to produce it and the parameters that can be used to describe it. The personnel working in and around the rig, especially at positions of highest impact should be properly protected with appropriate hearing protection plugs. As SPL levels near specific equipment exceed 100 db, it is important to ensure that workers wear additional protection and adhere to company and OSH standards at all times. This may warrant educating the workers on the dangers of noise impact and ensuring they understand the conditions and levels they are been exposed to. Noise barriers may also be utilized to protect those living and working close and around a drilling rig. These can be constructed out of inexpensive and readily available materials such as hay bales, which have been shown to be effective at lowering SLP levels. Further work into the design and proper setup of these walls is needed to fully understand how to best put them to use. ACKNOWLEDGEMENTS This study was kindly sponsored and hosted by Contact Energy Ltd. at the Wairakei Power Plant and adjacent steam field locations. The authors would like to thank Mark Green, Karen Heffer, Ralf Winmill and Mike Dunstall for their support and invaluable feedback in this study. REFERENCES [1] McGraw-Hill Concise Encyclopaedia of Physics. McGraw-Hill, [2] Auckland Geothermal Institute. Geothermal Noise and Silencer Design Course Notes [3] Australia/New Zealand Standards. Occupational noise management. Part 1: Measurement and assessment of noise immission and exposure [4] Morfey, C. L. Dictionary of Acoustics. Academic Press, [5] Bies D. A., Hansen, C. Engineering Noise Control: Theory and Practice. Taylor & Francis, [6] Kryter, K. D. The Handbook of Hearing and the Effects of Noise. Academic Press, [7] Cheremisinoff, N.P. Noise control in industry a practical guide. Noyes Publications, [8] New Zealand Department of Labour. Health & Safety Site. < [9] NTI Audio: Minilyzer ML1. < />. [10] NTI Audio: MiniSPL. < Minstruments/MiniSPL/>. [11] CASELLA CEL Inc.: CEL 460. < [12] Garmin Inc. GPS 60. <buy.garmin.com/shop/shop.do?pid=6446&ra=true>. [13] Pyrotek Inc. Wave Bar Sheets. <

6 Figure 3: a. Time-averaged vs. peak SPL; b. Exposure times above 70, 85, 90 and 100 db. Figure 4: Noise measurements behind each barrier for the three positions of the noise source. Figure 5: The effects of varying the angle of attack for the two Wave Bar barriers at 90 o & 66 o. Figure 6: Effects of wind interference on SPL measurements.